A look at how NDVI can aid cotton market participants’ trading and risk management strategies
By late May, before US cotton planting is completed, plant-health measurements can provide directionality on end-of-season US cotton abandoned acres and therefore prices, according to a Gro analysis of 20 years of Texas’ cotton abandoned acres data and Gro’s Normalized Difference Vegetation Index (NDVI) for the state.
The US is the third-largest cotton producer globally, and this year Texas planted nearly 60% of the US crop. Typically, farmers abandon acres when it is uneconomic to harvest a crop. Abandonment is much more prevalent for cotton than for other crops; historically, abandonment has had a substantial impact on final cotton production numbers.
Our analysis also shows that NDVI can aid cotton market participants’ trading and risk management strategies during the season because the crop’s vegetative health signals become more statistically significant as an indicator as the growing season progresses.
NDVI is a satellite-derived measure of plant health and greenness that can detect changes in crop conditions. It gauges plant health on a scale of 0-1, with 1 representing the healthiest vegetation. For most of this year, Texas’ cotton growing areas have had NDVI readings at the low end of the range, with levels ranging from 0.25 to 0.40, as seen in this display from Gro’s Climate Risk Navigator, weighted for acres planted to cotton in the state.
For a more complete description of NDVI, read this Gro report.
According to our analysis, for the last 20 years there has been a significant correlation between the average monthly NDVI level for each of May, June, and July in Texas’ cotton-growing areas and the state’s end-of-season abandoned cotton acres.
On average, the R-squared - a measure of how much of the variation in the dependent variable can be explained by the independent variable - between NDVI in Texas’ cotton-growing areas for the May through July period and the state’s abandoned cotton acres has strengthened from 0.54 in May to 0.77 in June to 0.86 in July.
Based on this historical relationship, NDVI at the end of May 2022 in Texas was pointing to at least a 42% Texas cotton abandonment rate for this year, compared to an average of 26% over the past 20 years. For June and July, it was pointing to an abandonment rate of 49.8% and 60.4%, respectively.
Gro’s forecasts using NDVI preceded by several months the USDA’s estimates for abandoned cotton acres. The USDA’s current estimated abandonment rate, made in its September WASDE report, is 68%. Due to this year’s drought in Texas, the USDA sharply cut its harvested acre estimate despite an increase in planted area. The September WASDE estimate for total US abandoned acres was 5.9 million, driven almost entirely by Texas’ 5.4 million abandoned acres.
In addition to serving as a leading indicator for abandoned US cotton acres, the NDVI can also be used to discern when market prices are not fully reflecting changes in crop conditions.
For example, for almost all of 2022, NDVI, weighted to Texas’ cotton-growing areas using Gro’s Climate Risk Navigator for Agriculture, has closely tracked levels seen in 2011, an anomalous year in which US cotton yields hit one of the lowest levels in the past 20 years. At the start of this year’s cotton planting season in early March, NDVI in Texas’ cotton-growing areas improved slightly. But buying December cotton futures to cover the crop year for $1.12 later that month would have made sense as NDVI at the start of the planting season closely mirrored 2011.
In early May and again in early June, NDVI for Texas cotton-growing areas showed some modest improvement, but both times the index remained below its 20-year NDVI average, suggesting that the crop’s health was still fragile. Selling into those improvements could have resulted in a 9%-11% profit, or a savings of $0.11 to $0.125 per pound. In its August WASDE report, the USDA cut cotton production estimates by 19% and cotton prices traded nearly 8% higher over the next two trading days.
Additionally, a profitable long position in cotton futures could have been reasonably reestablished in mid-July on the basis of NDVI once again dipping below 2011 levels.
Using a combination of Gro’s Portal and Climate Risk Navigator for Agriculture commodity investors can add to their understanding of crop health and, by extension, supply and price. In these examples (above), it would have been possible to identify the risk to the cotton crop in a key production region and capitalize on the trading opportunity.